Revert "drm/amd/display: fix bandwidth validation failure on DCN 2.1"
Subsystems (3)
PATCH 1/1 Revert "drm/amd/display: fix bandwidth validation failure on DCN 2.1"
From: Ivan Lipski <[email protected]>
This commit causes dmesg-warn on several IGT tests on DCN 3.1.6:
*ERROR* link_enc_cfg_validate: Invalid link encoder assignments - 0x1c
Affected IGT tests include:
amdgpu/[amd_assr|amd_plane|amd_hotplug]kms_atomickms_colorkms_flipkms_propertieskms_universal_plane
and some other tests
This reverts commit b7ebd39e2922f642c7ee63ade4a4a5a1ef675d84.
--- drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +1 -1
@@ -10752,7 +10752,7 @@ static int amdgpu_dm_atomic_check(struct drm_device *dev,DRM_DEBUG_DRIVER ("drm_dp_mst_atomic_check() failed\n" ); goto fail; } - status =dc_validate_global_state (dc, dm_state->context, false); + status =dc_validate_global_state (dc, dm_state->context, true);if (status !=DC_OK ) {DRM_DEBUG_DRIVER ("DC global validation failure: %s (%d)" ,dc_status_to_str (status), status);
Signature · 1 line
2.34.1
[AMD Official Use Only - General]
@Siqueira, Rodrigo<mailto:[email protected]>
________________________________
From: LIPSKI, IVAN <[email protected]>
Sent: January 5, 2024 7:40 PM
To: [email protected] <[email protected]>
Cc: [email protected] <[email protected]>; Choi, Nicholas <[email protected]>; Deucher, Alexander <[email protected]>; Koenig, Christian <[email protected]>; Wentland, Harry <[email protected]>; LIPSKI, IVAN <[email protected]>; Melissa Wen <[email protected]>; Mahfooz, Hamza <[email protected]>
Subject: [PATCH 1/1] Revert "drm/amd/display: fix bandwidth validation failure on DCN 2.1"
From: Ivan Lipski <[email protected]>
This commit causes dmesg-warn on several IGT tests on DCN 3.1.6:
*ERROR* link_enc_cfg_validate: Invalid link encoder assignments - 0x1c
Affected IGT tests include:
amdgpu/[amd_assr|amd_plane|amd_hotplug]kms_atomickms_colorkms_flipkms_propertieskms_universal_plane
and some other tests
This reverts commit b7ebd39e2922f642c7ee63ade4a4a5a1ef675d84.
--- drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +1 -1
@@ -10752,7 +10752,7 @@ static int amdgpu_dm_atomic_check(struct drm_device *dev,DRM_DEBUG_DRIVER ("drm_dp_mst_atomic_check() failed\n" ); goto fail; } - status =dc_validate_global_state (dc, dm_state->context, false); + status =dc_validate_global_state (dc, dm_state->context, true);if (status !=DC_OK ) {DRM_DEBUG_DRIVER ("DC global validation failure: %s (%d)" ,dc_status_to_str (status), status);
--
2.34.1
Quoted 23 lines · IVAN
[AMD Official Use Only - General]
@Siqueira, Rodrigo<mailto:[email protected]>
________________________________
From: LIPSKI, IVAN <[email protected]>
Sent: January 5, 2024 7:40 PM
To: [email protected] <[email protected]>
Cc: [email protected] <[email protected]>; Choi, Nicholas <[email protected]>; Deucher, Alexander <[email protected]>; Koenig, Christian <[email protected]>; Wentland, Harry <[email protected]>; LIPSKI, IVAN <[email protected]>; Melissa Wen <[email protected]>; Mahfooz, Hamza <[email protected]>
Subject: [PATCH 1/1] Revert "drm/amd/display: fix bandwidth validation failure on DCN 2.1"From: Ivan Lipski <[email protected]>
This commit causes dmesg-warn on several IGT tests on DCN 3.1.6:
*ERROR*link_enc_cfg_validate: Invalid link encoder assignments -0x1cAffected IGT tests include:
amdgpu/[amd_assr|amd_plane|amd_hotplug]kms_atomickms_colorkms_flipkms_propertieskms_universal_plane
Quoted 3 lines
and some other tests
This reverts commit
b7ebd39e2922f642c7ee63ade4a4a5a1ef675d84.
I'm not opposed to reverting this in the short term, but I don't see the connection between doing a full validation and link encoder assignment errors. It seems more like the change unveiled an underlying issue rather than causing the error. I don't see those errors on DCN 2.1 and 3.0.1.
Unfortunately, I don't have a DCN 3.1.6 for debugging :( It'd deserve further investigation.
Melissa
Quoted 26 lines
Cc Melissa Wen <[email protected]>Cc Hamza Mahfooz <[email protected]>Signed-off-by Ivan Lipski <[email protected]>--- drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +1
-2@@ -10752,7 +10752,7 @@ static int amdgpu_dm_atomic_check(struct drm_device *dev,DRM_DEBUG_DRIVER ("drm_dp_mst_atomic_check() failed\n" ); goto fail; } - status =dc_validate_global_state (dc, dm_state->context, false); + status =dc_validate_global_state (dc, dm_state->context, true);if (status !=DC_OK ) {DRM_DEBUG_DRIVER ("DC global validation failure: %s (%d)" ,dc_status_to_str (status), status); --2.34.1
Rodrigo Siqueira Jordao
Reviewed-byQuoted 26 lines · Melissa Wen
Quoted 25 lines · IVAN
[AMD Official Use Only - General]
@Siqueira, Rodrigo<mailto:[email protected]>
________________________________
From: LIPSKI, IVAN <[email protected]>
Sent: January 5, 2024 7:40 PM
To: [email protected] <[email protected]>
Cc: [email protected] <[email protected]>; Choi, Nicholas <[email protected]>; Deucher, Alexander <[email protected]>; Koenig, Christian <[email protected]>; Wentland, Harry <[email protected]>; LIPSKI, IVAN <[email protected]>; Melissa Wen <[email protected]>; Mahfooz, Hamza <[email protected]>
Subject: [PATCH 1/1] Revert "drm/amd/display: fix bandwidth validation failure on DCN 2.1"From: Ivan Lipski <[email protected]>
This commit causes dmesg-warn on several IGT tests on DCN 3.1.6:
*ERROR*link_enc_cfg_validate: Invalid link encoder assignments -0x1cAffected IGT tests include:
amdgpu/[amd_assr|amd_plane|amd_hotplug]kms_atomickms_colorkms_flipkms_propertieskms_universal_planeand some other tests
Quoted 7 lines
Quoted 1 line
This reverts commit
b7ebd39e2922f642c7ee63ade4a4a5a1ef675d84.I'm not opposed to reverting this in the short term, but I don't see the connection between doing a full validation and link encoder assignment errors. It seems more like the change unveiled an underlying issue rather than causing the error. I don't see those errors on DCN 2.1 and 3.0.1.
Yeah... I totally agree with you. I'll try to get one DCN 316 device to validate this issue. For now, let's revert this change since it causes a large regression in our CI.
Also applied to asdn.
Thanks
Siqueira
Quoted 32 lines
Unfortunately, I don't have a DCN 3.1.6 for debugging :( It'd deserve further investigation.
Melissa
Quoted 26 lines
Cc Melissa Wen <[email protected]>Cc Hamza Mahfooz <[email protected]>Signed-off-by Ivan Lipski <[email protected]>--- drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +1
-2@@ -10752,7 +10752,7 @@ static int amdgpu_dm_atomic_check(struct drm_device *dev,DRM_DEBUG_DRIVER ("drm_dp_mst_atomic_check() failed\n" ); goto fail; } - status =dc_validate_global_state (dc, dm_state->context, false); + status =dc_validate_global_state (dc, dm_state->context, true);if (status !=DC_OK ) {DRM_DEBUG_DRIVER ("DC global validation failure: %s (%d)" ,dc_status_to_str (status), status); --2.34.1